JP2001526397A - How to predict the presence of curves in road sections - Google Patents

Abstract

(57) [Summary] The present invention is arranged on a vehicle which transmits radar signals and records the reception reflection thereof to provide information indicating the position of an object group placed in an area in front of the vehicle. A method and apparatus for predicting the presence of a curve in a road section ahead of a vehicle in connection with the use of radar. According to the invention, a value representing the direction of the virtual line group between the object groups is derived, and corresponding values, each defining a position of each one of the line groups between the object groups, are derived from the direction. You. Thereafter, it is determined whether at least a subset of the values of direction and corresponding position satisfy a functional relationship between direction and position, whereby the presence of the curve is predicted based on them.

Description

DETAILED DESCRIPTION OF THE INVENTION

FIELD OF THE INVENTION The present invention relates to the use of radar located on a vehicle to transmit radar signals and record the received reflections to determine the location of objects in the area in front of the vehicle. The present invention relates to a method for predicting the presence of a curve in a road portion in front of a vehicle, which provides the information shown. The invention also relates to an apparatus for performing the method.

[0002] type of background the invention described above, for example, a vehicle mounted radar system used cars, buses, trucks, and etc., now in course of development. The purpose of the system is mainly based on the detection of vehicles, stationary obstacles and similar objects in front of the vehicle, such as cruise control and collision warning, but also in the event of an emergency collision in the future And assisting the vehicle driver by providing functions such as automatic braking or other types of evasion maneuvers.

[0003] The radar system provides a prediction of the direction of the course ahead of the vehicle based on the detected objects in the area in front of the vehicle, ie whether the road is straight or curved and in the latter case. It is often required to determine the size of an existing curve. Such objects include, in addition to vehicles traveling in front of the vehicle, road signs, card rails, and the like, which are used to predict course directions in this regard.
Etc. may be included.

One method for determining the curvature of a road is known from EP 464 821, in which the virtual position of the road is adjusted to correspond to the position of the object appearing in the image to the maximum possible extent. A function approximation method is used to create a virtual function relationship that describes the state of the target curvature.

Objects of the invention For example, a problem with the approximation method of the prior art scheme described above is that reflections originating from objects that do not limit the direction of travel affect the approximation of the functional relationship, that is,
Any object that does not delimit or limit the course contributes to disturbing the appearance of the function, and thus causes false prediction of the course direction.

[0006] In addition, if the object is assumed to originate from several lines or curves, approximations of some functions will be required. In this regard, it will be seen that the required processor capacity of the system increases with the number of lines it must be able to determine.

It is, therefore, an object of the present invention to reduce or eliminate the problems associated with the effects of unrelated objects in predicting the course ahead of a vehicle.

Another object of the present invention is to simplify the arithmetic processing when it is expected that some lines or curves will be found in predicting the course direction ahead of the vehicle.

It is yet another object of the present invention to provide a solution that reduces the computational capacity required for a radar system of the type described above.

SUMMARY OF THE INVENTION The above objects are achieved by the invention as defined in the appended claims.

According to a first aspect of the present invention, there is provided a method of the kind described as an introduction and characterized by the following steps: deriving a value representing the direction of a virtual line between the objects and each of them. Associating a derived corresponding value with the direction, wherein the value defines a position of each one of the lines between the objects; and at least a subset of the values for direction and corresponding position is a functional relationship between direction and position. And predicting the presence of said curvature based thereon.

Thus, rather than starting from the actual position of the objects, the present invention starts from the directions indicated by the objects, more specifically the directions of the virtual lines between the objects, and in those directions. It is based on the idea of laying the foundation for course direction prediction. Conventional analytic methods are not required to determine these directions since their directions relate to virtual lines between virtual objects. Rather, their directions can be derived directly from information about the position of each combination of the two objects, which requires less processor capacity.

The guidance of the direction of a virtual line between objects in a radar image is similar to some kind of spatial discrimination of position information. As a general rule, this means that the information about the horizontal position of the object is modified. However, this is because the direction values of lines between objects at the same distance from the vehicle but at different lateral positions, for example, reflections from the guardrails on both sides of the road, are mutually enhanced in subsequent analyses. It has the advantage of fitting. When a known curvature approximation based directly on the position of the object is used, individual curvature approximations must be performed for the different possible curvatures in the image. According to the present invention, the curves on both sides of the road together give directions which jointly indicate the course directions, at least initially without the need for separate calculations.

According to a preferred embodiment of the present invention, based on a number pairs obtained from said values of the directions and the corresponding positions, the functional relationship between the directions and the positions is satisfied with at least a subset of said number pairs. The Hough transform is used to determine whether and, if so, to determine the possible (probable) occurrence parameters of this functional relationship. In this context, it should be noted that the use of the Hough transform, which was originally associated with image analysis to detect the groups of lines there, is well known to those skilled in image processing. Note that the Hough transform according to the invention is performed on a direction / position space basis and not on the original space space.

If, for example, the course ahead of the vehicle can be simulated as an arc, the relationship between direction and distance will be essentially linear. In this case, the Hough transform is used to find the aforementioned linear relationship between direction and distance for at least a subset of several pairs obtained from the above values of direction and corresponding position.

More specifically, the Hough transform preferably transforms several pairs, obtained from the values of direction and corresponding position, into a functional relationship of said direction whose dimensions are a function of position with respect to a virtual range. A step of converting into a curve in a parameter space corresponding to a parameter group to be included and a combination of parameters obtained from the curve indicating a likely appearance of the functional relationship; one between the curves in the parameter space Determining the intersections, and calculating a combination of parameters indicating the specific appearance of the functional relationship corresponding to the specific appearance of the curvature existing on the road portion based on the intersections.

It will be seen that the intersection in the parameter space limits the parameters of the functional relationship between direction and position, and in turn the relationship between direction and position limits the curvature of the road. .

Since each number pair is the source of an individual curve in the parameter space, a set of number pairs is the source of a set of curves in the parameter space, so that multiple intersections exist between these curves. You will see that it will. However, this does not mean that every intersection corresponds to a related functional relationship. Only if the intersections are gathered in a small area of the parameter space, does this reach the conclusion that there is a common functional relationship for several pairs, and only then is the course prediction considered to be present.
Of course, many false intersections will appear, but the probability that they will combine to create a significant false peak is small.

It is desired that the Hough transform is based on a linear relationship expressed as a first-order polynomial, and that the parameter space is two-dimensional. However, the transform can also be used for multidimensional or non-linear relationships. For example, the above-mentioned curves in the parameter space may be straight lines and sinusoidal relationships or higher order polynomials.

The analysis of the curve in the parameter space created by the Hough transform is advantageously performed by increasing or otherwise changing the element values in a matrix having the same dimensions as the parameter space. When the parameter space is represented by a matrix, to determine the relevant intersections in the parameter space,
Initial mean value formation, whose purpose is to flatten out the amplitudes of the matrix, is advantageously used, followed by the determination of the local maximum of the matrix. You. By choosing an appropriate solution of the matrix, ie the number of elements in the matrix, for the display of the width of the curves (calculated as the number of elements in the matrix) and of the size of the window used to form the average value , It is ensured that the relevant local maximum is derived from the matrix.

First, all elements in the matrix are set to zero. Next, the matrix elements corresponding to the curve groups generated by the several pairs of Hough transforms are updated. This means that each number pair results in an increment of all elements on the surface of a lower dimension than the dimension of the matrix, ie the curve in the case of a two-dimensional matrix.

For example, a major advantage of using the Hough transform over least-squares approximation is that several pairs that do not represent the curvature of the road do not affect the prediction of the course direction, and that the parameter space Is that it is not necessary to determine some of the curves being searched until the actual conversion to has been accomplished, or to determine which few pairs belong to which curve.

For a more detailed description of the Hough transform in image processing, see the book “Dig by Rafael C. Gonzales and Richard E. Woods, published by Addison-Wesley publisher.
ital Image Processing ".

Although utilizing the Hough transform is a preferred alternative for analyzing the direction and position induced according to the present invention, it is clear that other methods can be applied for this purpose. Would. According to an alternative embodiment, a least-squares approximation of the functional relationship between the direction and the position of the virtual road part is used for several pairs resulting from said values of the direction and the corresponding position, after which the likely curvature present in the road part Is determined based on the least squares approximated functional relationship described above.

Advantageously, in addition to determining the direction of lines between objects in the radar image,
The direction of the moving object is also determined for the object determined to be the moving object based on the radar information collected in two separate cases. By deriving a corresponding value representing the position of. Thus, several pairs used in subsequent analysis, preferably implemented using the Hough transform, relate to lines between objects in the image as well as the direction of motion of the moving objects. It will be appreciated that the direction of movement of the vehicle in question may also be used as a basis for a corresponding few pairs.

To the extent that a moving object is processed differently than a stationary object, the guidance of the direction of the line between the different objects is such that the line between the stationary object and the moving object is likely to represent the course direction. Therefore, it is preferably limited to lines between stationary objects.

Furthermore, the analysis according to the invention is preferably limited to include only guidance direction values that do not deviate from the running direction of the vehicle as much as a preset value. This is based on the assumption that a direction that extends somewhat laterally from the direction of travel of the vehicle is unlikely to represent the road on which the vehicle is traveling. Thus, if such apparently false direction values were easily ignored, the analysis would be faster and easier.

In addition, the analysis according to the invention is preferably limited to include only direction values that do not deviate from the previously predicted course direction at the corresponding position as much as the preset values. This is because the direction of the road far away from the vehicle in question can be significantly different from the direction of travel of the vehicle, and the resulting categorization of the relevance of the direction values is preferably a prediction at the location in question. It is based on the recognition that the course direction should be related to the extent that the prediction was achieved in a previous phase.

For example, the direction of motion of a vehicle traveling in front of the vehicle in question probably corresponds to the course direction, while the direction of the individual line between any two stationary objects is Since there is no clear relationship with the direction, it is possible to achieve a more reliable analysis by weighting the corresponding number pairs separately, depending on the classification of the object or group of objects from which the number pairs originate. It is possible. For example, several pairs corresponding to the position and direction of motion of the moving object are preferably weighted heavier than several pairs corresponding to the direction and position of the individual line between the two objects. Yet another alternative is to weight several pairs of moving objects separately depending on their speed.

To further simplify the process, it is sufficient to form lines only between certain objects rather than between all found objects. For example, it is very unlikely that a line between two objects placed far apart from each other will actually correspond to the course direction. Road-defining objects, such as guardrails, utility poles, etc., usually appear relatively regularly and at short intervals, so that lines between objects at such short intervals from each other may actually It is more likely to represent the course direction. Therefore,
In the preferred embodiment, only lines between objects located less than a preset distance from each other, for example, 10 meters, are processed.

In connection with the continuous processing of the radar information recorded at successive times, it is appropriate to temporarily store the values from the previous processing, for example by storing the element values of the matrix. It is. During subsequent processing, it is sufficient to derive new direction and position values for the line group to / from the new object in relation to the previous information then the new values are used for updating the matrix. When utilizing the aforementioned updates, it is important to take into account the altered distance resulting from the movement of the vehicle in question.

The processing of the guidance values described above may be conveniently divided into partial processing operations on individual sub-areas of the area in front of the vehicle. The aforementioned sub-areas can be defined in relation to the vehicle or in relation to its environment. Thus, the choice of subarea definition will affect the choice of how previously derived number pairs or calculated matrices are stored and updated.

Each resulting direction, represented by a respective line or direction of movement, should be considered with respect to the position where the direction is believed to be related to that position. Thus, according to the invention, a corresponding position value is derived for each direction value obtained. These locations may be represented in several different ways. A preferred alternative would be for the position to be expressed as the distance therefrom of the vehicle in question. According to another alternative, the position is represented as a distance from a selected reference point, which may be defined in relation to the vehicle or in relation to its environment. For example, one of the objects may be selected as a reference point.

The position value may be, for example, a linear distance between the vehicle and the line / object, a distance between the vehicle and the line / object projected on an axis extending along the traveling direction of the vehicle in question, or a distance between the vehicle and the line / object. It may be expressed as an arcuate (arc) distance to the line / object, as seen from the vehicle, taking into account the angle up to it. If the processing is performed in sub-areas, as described above, the location information may come from knowledge of which sub-area is associated with a particular line / object. Similarly, directions are preferably indicated as spatial directions or angles in a coordinate system defined for the vehicle, but of course, other methods of defining directions are also possible, e.g., with a ground-based baseline.
sed) It may be used in connection with a coordinate system. Further, it will be appreciated that all direction values may preferably be related to the direction of travel of the vehicle in some manner.

For example, when determining the distance to the direction of a line, the distance to the midpoint between the two objects from which the line originates is that the direction of each line is probably the best at the intermediate distance between the objects. It is advantageously chosen because it represents. It should be noted, however, that the invention is, of course, not limited to this choice, and that other foundations may be used to determine the distance between the vehicle in question and the line between the two objects. It is. For example, the distance from the vehicle to the closer of the two objects can be selected.

According to yet another embodiment of the invention, the initial derivation of several pairs is repeated for the actual few pairs themselves. By considering the few pairs as points in a two-dimensional image, a directional analysis is performed at yet another time on the lines between these points (several pairs) before making the actual determination of the functional relationship. You can also. In this case, the result can be regarded as a secondary guidance of the initial position information.

The present invention is advantageously implemented using conventional microprocessor technology, and according to a second aspect, the present invention is therefore directed to the steps and measurements set forth in the appended claims and the above discussion. It will be appreciated that the present invention pertains to devices, such as microprocessors, for performing the following.

Further features and advantages of the invention are set forth in the following preferred embodiments of the invention, as well as in the appended claims. Embodiments of the present invention will now be described, by way of example, with reference to the accompanying drawings:

Detailed Description of the Preferred Embodiment For simplicity, FIG. 1 shows a schematic diagram of the position of an object detected by a vehicle-mounted radar in an area located in front of the vehicle. The radar provides continuously updated position information, for example, ten times per second, by transmitting a scanning radar signal and recording its received reflections from objects located in the front area of the vehicle. Although the description related to the figures in this specification is based on the application of the present invention to any amount of information, the present invention, when combined, provides better decision support data. The information obtained continuously can also be used to advantage in practice.

In FIG. 1, the radar is mounted on the vehicle in question. In the illustration, the radar is schematically indicated by reference numeral 20 in the figure, and the positional relationship of the vehicle 20 with respect to other objects is schematically illustrated.

As can be seen from FIG. 1, a number of objects 31, 32, 33 are arranged in front of the vehicle. One of the objects is a vehicle 31 traveling in the opposite direction of the vehicle in question. Another object is a vehicle 32 that is traveling in the same direction as the vehicle in question and is located on a road curve. Other objects 33 are stationary objects in the environment, such as trees, utility poles, guardrails, road signs, and the like. According to the prior art, moving objects are distinguished from stationary objects, for example by comparing successively performed scans or by analyzing the frequency content of the individual reflections.

As can be seen from FIG. 1, predicting the direction of the path in front of the vehicle, schematically shown by the dashed line 40 in the figure, depends on the position of the stationary object 33 and the positions of the moving objects 31 and 32 and their positions. This is possible based on the direction of movement.

FIGS. 2, 3 and 3b divide the shape of the road into straight lines and curves in the class of predictions. Most roads, especially those built for high speeds over 60 km / h, consist of straight lines, circles, and segments or sections that can be approximated with high accuracy by clothoids (straight lines are infinite Note that this can be viewed as a circle having a radius of curvature of.

In the case of a straight line, the direction of the line is constant (as in road parts I and V in FIG. 2). In this case, as shown by I and V in FIGS. 3a and 3b, with respect to the traveling distance (range), the derivative of the direction Dir of the passage is zero, and with respect to the traveling distance, the second derivative of the direction of the passage is also zero. is there.

In the case of a circle, the curvature of the line is constant (as in road part III of FIG. 2). In this case, as shown by III in FIGS. 3a and 3b, the derivative of the direction Dir of the path with respect to the mileage (range) is constant, and the second derivative of the direction of the path with respect to the mileage is therefore zero. is there.

Clothoids are defined by the fact that the radius of curvature R of a line is proportional (or inversely proportional) to the length of the arc. Thus, as shown by II and IV in FIGS. 3a and 3b, the derivative of the direction of the path as a function of the mileage is proportional to the mileage,
In addition, the second derivative in the direction of the passage with respect to the traveling distance is constant.

It will be seen that by looking for a straight line in a diagram of the type shown in FIGS. 3a and 3b, it is possible to establish the presence of a road curvature. Some of that will be discussed in more detail below with reference to FIGS. 7a and 7b, among others.

FIG. 4 illustrates a schematic selection of lines and positions (distances) of the object of FIG. 1, according to an embodiment of the present invention. According to the invention, a virtual line between the stationary objects 33 is calculated.
In this embodiment, the calculation of the line from all objects to all objects is not performed, but rather only the line between each stationary object and a stationary object placed at a fixed distance therefrom, for example, within 10 meters. . This choice is based on the assumption that lines between objects placed far away from each other are less likely to represent the actual direction of the path. In the basic case, the two objects are connected by only one line, ie no line is counted twice.

Each line represents a direction. This direction can be indicated in several ways. For example, as a quotient y / x shown at 53 or as an angle α shown at 54. Note that these two direction indications can be related to the direction of travel of the vehicle 20 (△ y / △ x = ∞ or α = 90 °). The direction and position in FIG.
Also note that it is displayed with respect to the coordinate system defined for. However, it will be appreciated that these values can be calculated, just as well, based on a coordinate system based on the ground.

In the case of the moving objects 31 and 32 as well, as indicated by the arrows adjacent to the vehicles 31 and 32, the direction of movement of the object itself is shown in relation to the traveling direction of the vehicle 20 by the same direction display method. It is. This usually means that it is necessary to compare the position of the moving object at two different times.

According to yet another embodiment, a certain set limit value, for example, 50
Any guidance direction that deviates by more than a degree is simply ignored. This is because such directions are unlikely to represent the direction of the passage ahead of the vehicle. Limit values for deviations related to the predicted direction of the path based on previous predictions made with a certain degree of reliability may also be used.

Each acquisition direction, represented by a respective line or direction of motion, applies with respect to its position. Accordingly, for each obtained direction value, a corresponding position value is derived, which is displayed on the vehicle as a reference point in FIG. This distance value can be displayed in several ways. For example, as indicated by the dashed line 50,
As the shortest distance between the vehicle 20 and the line / object, or as indicated by the dashed line 51, as the distance between the vehicle and the line / object projected on an axis extending along the direction of travel of the vehicle, or as a dashed line 52 As shown in, as seen from the vehicle, as an arc-shaped distance to the line / object, taking into account the angle to it. For the lines found, the midpoint of each line is beneficially selected when determining its position as described above. This is because the direction of each line probably best describes the direction of the path at intermediate distances between objects.

As described above, if the directions and distances derived from FIG. 4 are depicted as points in a diagram showing directions as a function of distance, a relationship similar to that shown in FIG. 5 is obtained. This figure may be said to represent the spatial derivative of the initial position information, as discussed with reference to FIGS. 2, 3a and 3b above, so the existence of an essentially circular curve in spatial space is As indicated by the dashed line in FIG. 5, the data is displayed in an essentially linear relationship between the point groups in the direction / distance diagram. By determining the parameters of the straight line based on the obtained point group, it is easy to calculate the parameters of the corresponding curve on the road.

A preferred method for deriving the parameters of the straight line represented between the point clouds of FIG. 5 using the Hough transform will now be described with reference to FIGS. 6a, 6b, 7a and 7b.

As shown in FIG. 6a, suppose we want to determine a straight line between two points (or several pairs) (x ₁ , y ₁ ) and (x ₂ , y ₂ ) of the image with respect to axes x and y. In this case, the line determined in the xy plane can be described by the equation y = ax + b, where a and b are parameters that must be determined for the line to be clearly defined . Since the obtained line intersects the point (x ₁ , y ₁ ), the natural consequence is that y ₁ = ax ₁ +
becomes b. This relationship can also be expressed as b = y ₁ −ax ₁ , which, as shown in FIG. 6b, has a surface with parameters a and b as its axis, the so-called (in this case, two-dimensional ) In the parameter space. This straight line in the parameter space represents all combinations of a and b that are a group of lines in the xy plane that intersects the point (x ₁ , y ₁ ).

The line to be obtained is also a point (x₂, Y₂), So of course, y₂ = ax₂ + B. Correspondingly, this is b = y₂−ax₂Which can also be expressed as
It can be regarded as a straight line in the data space. Therefore, the parameter space in FIG.
Each point on the xy plane in FIG. Both points (x ₁ , Y₁) And (x₂, Y₂) Are the two pairs in parameter space.
Parameter found in the response line, i.e., as indicated by the dashed line in FIG.
With parameters a and b corresponding to the intersections between lines in data space
. By determining the intersection in the parameter space of FIG. 6b, the line in the xy plane of FIG. 6a is clearly determined.

The description corresponding to that given above with reference to FIGS. 6 a and 6 b then corresponds, in general, to the diagram described with reference to FIG. A description will be given based on a diagram showing directions as a function of distance. For each of the six points in FIG. 7a, a corresponding line in the parameter space ab is determined. In practice, this line is represented by a weighted element of the parameter matrix. In the case of five of these lines, it is possible to determine the likely intersections that will give the corresponding line in the direction / distance diagram. Since the lines do not exactly intersect the same point, a likely common intersection is the local maximum in the aforementioned matrix following preliminary mean value formation or its equalization. Is derived by calculating The maximum value thus determined in the parameter space can then be transformed into a line in the direction / distance diagram, which can then be transformed into a curve in the first spatial space.

The sixth line, corresponding to the divergence point in the direction / distance diagram, is plotted at a point cloud far away from the other intersections after the averaging and calculation of the maxima described above.
It intersects the other group of lines in 7b, because it is assumed to correspond to a likely line in the direction / distance plane. Thus, the divergent point in FIG. 7a does not affect the determination of the parameter values of the functional relationship, as opposed to the case where the line in direction / metric space is determined, for example, by a least squares approximation based on all points.

It will be appreciated that the steps for practicing the invention do not include the step of actually deriving the values obtained for the directions and positions (distances) in figures of the type shown in FIGS. 5 and 7a. Rather, the corresponding curves can be formed in a parameter space matrix (corresponding to FIG. 7b), starting directly from the directions (several pairs) obtained in the radar image in FIG. Accordingly, it will also be appreciated that the location information need not be displayed in the form of an image of the type shown in FIG. 1 to perform the steps of the invention. Rather, the computation can be performed based on raw signal data without intermediate image analysis.

FIG. 8 illustrates the division of a region in front of the vehicle into sub-regions (sub-areas) according to an embodiment of the invention. In FIG. 8, region A corresponds to the first portion closest to the vehicle and region B corresponds to an adjacent portion that is farther away from the vehicle. For example, in the phase schematically illustrated in FIG. 8, there will be virtually no lines in region A, while in region B curvature will likely be predicted.

As a result of this division, the arithmetic processing described with reference to the above figure, advantageously
Performed separately for isolation regions. In this regard, the Hough transform described above is preferably implemented with separate matrices for separation areas A and B. Regions A and B are vehicles
It can be defined in terms of 20 or its environment. In the latter case, regions A and B will move toward the vehicle as the vehicle moves forward in its environment, and the new regions will replace regions A and B one after the other.

Referring to the illustration above, all possible objects representing all existing radar reflections need not be considered in the computation. Rather, the operation may be advantageously limited to reflections that can be determined with some confidence to correspond to a real object.

The above description of the preferred embodiments is provided by way of example only, and modifications and combinations thereof may be practiced within the scope of the invention, which is limited by the appended claims. It will be appreciated that gains are obtained.

[Brief description of the drawings]

FIG. 1 shows a schematic diagram of the position of a group of objects associated with a vehicle.

FIG. 2 schematically shows the geographical division of roads into predefined straight lines and curves.

3a and 3b show diagrammatically the first and second derivatives of the direction of the road as a function of the range of FIG.

FIG. 4 schematically illustrates the selection of lines and distances for the objects of FIG. 1 according to an embodiment of the present invention.

FIG. 5 schematically shows a diagram of the direction as a function of the distance from the radar image in FIG.

FIGS. 6a and 6b show examples of using the Hough transform for line detection in the xy plane.

7a and 7b schematically show an example of using the Hough transform for line detection in the diagram shown in FIG. 5;

FIG. 8 schematically illustrates the division of the processing of the group of objects of FIG. 1 into individual regions in accordance with an embodiment of the present invention.

[Procedural Amendment] Submission of translation of Article 34 Amendment of the Patent Cooperation Treaty

[Submission date] June 17, 1999 (June 17, 1999)

[Procedure amendment 1]

[Document name to be amended] Statement

[Correction target item name] Claims

[Correction method] Change

[Correction contents]

[Claims]

──────────────────────────────────────────────────続 き Continuation of front page (81) Designated country EP (AT, BE, CH, CY, DE, DK, ES, FI, FR, GB, GR, IE, IT, LU, MC, NL, PT, SE ), OA (BF, BJ, CF, CG, CI, CM, GA, GN, GW, ML, MR, NE, SN, TD, TG), AP (GH, GM, KE, LS, MW, SD, SZ, UG, ZW), EA (AM, AZ, BY, KG, KZ, MD, RU, TJ, TM), AL, AM, AT, AU, AZ, BA, BB, BG, BR, BY, CA, CH, CN, CU, CZ, DE, DK, EE, ES, FI, GB, GD, GE, GH, GM, HR, HU, ID, IL, IN, IS, JP, KE , KG, KP, KR, KZ, LC, LK, LR, LS, LT, LU, LV, MD, MG, MK, MN, MW, MX, NO, NZ, PL, PT, RO, RU, SD, SE, SG, SI, SK, SL, TJ, TM, TR, TT, UA, UG, US, UZ, VN, YU, ZWF terms (reference) 5H180 AA01 CC14 LL01 LL04 LL09 LL15 5J070 AC01 AE01 AF03 AL02 BF02 BF19

Claims (27)

[Claims] 1. Related to the use of radar located on a vehicle to transmit radar signals and record their received reflections to provide information on the position of a group of objects located in an area in front of the vehicle. A method for predicting the presence of a curve in a road portion in front of a vehicle: derives a value representing the direction of a virtual line group between said object groups, and each position of one of said line groups between said object groups. Associating a derived corresponding value with said direction, defining whether at least a subset of said values of direction and corresponding position satisfy a functional relationship between direction and position, and based thereon, Predicting the presence of a curve. 2. The step of determining whether at least a subset of said values of direction and corresponding position satisfies a functional relationship between direction and position comprises a number of pairs of Hough transforms derived from said values of direction and corresponding position. The method according to claim 1, wherein the method is performed. 3. The step of determining whether at least a subset of said values of direction and corresponding position satisfies a functional relationship between direction and position, wherein the step of determining whether the value of direction and corresponding position is a curve in a number of pairs of parameter spaces obtained from said values of direction and corresponding position. Hough transform to a group, determining one or more intersections between curves in the parameter space, and determining values of the parameter group of the functional relationship based on these intersections. Item 1. The method according to Item 1. 4. The step of determining whether at least a subset of said values of direction and corresponding position satisfies a functional relationship between direction and position, comprising: determining a number of pairs obtained from said values of direction and corresponding position; Parameters in the parameter space where the dimensions correspond to a combination of parameters included in the functional relationship of the direction as a function of position with respect to the virtual range, a parameter obtained from the curve indicating a likely appearance of the functional relationship. Converting to a curve; and determining one or more intersections between the curves in the parameter space and identifying the functional relationship corresponding to a particular manifestation of the curve present on the road portion based on these. Calculating the combination of parameters indicating the appearance of the method. 4. The method according to claim 1, 2 or 3, . 5. The method according to claim 4, wherein the display of the curve in the parameter space is achieved by changing an element value in a matrix having the same dimensions as the parameter space. 6. The method of claim 5, wherein said determining one or more intersections comprises determining a maximum in said matrix. 7. The method according to claim 4, wherein the functional relation and the curve are obtained from a first-order polynomial. 8. The step of determining whether at least a subset of said values of direction and corresponding position satisfy a functional relationship between direction and position, comprising: Least-squares approximation of a functional relationship between direction and position for the road segment; and determining a likely occurrence of a curve in the road segment based on the least-squares approximated functional relationship. The method according to claim 1, wherein the method is performed. 9. The method according to claim 1, further comprising the step of processing only a guidance value in a determined direction that does not deviate from the traveling direction of the vehicle to a value that is equal to or smaller than a preset limit value. The method according to one. 10. A guide value in a determined direction which does not deviate from a predicted direction of a path at a position corresponding to each direction determined based on a previous prediction of a direction of a path to a degree that is equal to or greater than a preset limit value. 10. A method as claimed in any one of the preceding claims, comprising the step of processing only. 11. The method according to claim 1, further comprising the step of weighting the existence of said number pairs depending on the classification of the object or group of objects from which said number pairs originate. The method described in one. 12. The method of claim 1, wherein the step of deriving the value comprises measuring for at least one of the group of objects a value representative of a direction of a line between the object and another object. 12. The method according to any one of items 11 to 11. 13. The method of claim 1, wherein the measuring comprises deriving, for the at least one object, a plurality of values representing a direction of a group of lines between the object and a plurality of other objects. Item 13. The method according to Item 12. 14. The method according to claim 12, comprising performing the measurement on a plurality of objects. 15. The method according to claim 12, further comprising the step of performing said measurement on a group of objects as added to the objects already processed in a previous step.
15. The method according to claim 13, 13 or 14. 16. The method according to claim 1, wherein the value of the position of the line group is represented as a distance from a reference point to the line group. 17. The method according to claim 1, wherein the value of the position of the line group is represented as a distance from a vehicle to the line group. 18. The method according to claim 1, wherein the distance from the vehicle to each line is selected to be the distance from the vehicle to each point on each line.
17. The method according to 17. 19. The method according to claim 18, wherein each point is selected midway between a group of objects between which a corresponding line extends. 20. The method of claim 18, wherein the distance is approximated as a shortest distance between the vehicle and a point on each line. 21. The method according to claim 1, further comprising the step of guiding only the direction of a group of lines between a group of objects arranged within a predetermined distance from each other.
0. The method according to any one of 0. 22. The method as claimed in claim 1, wherein the value representing the direction corresponds to the direction of travel of the vehicle. 23. The method according to claim 1, wherein the direction is obtained with respect to a coordinate system defined with respect to the vehicle. 24. A method according to claim 1, wherein the direction is obtained in relation to a ground-based coordinate system. 25. The method according to claim 1, wherein the steps are performed in separate processing operations for different sub-areas of the region. 26. The step of deriving a value comprises deriving, for an object determined to be a moving object based on information obtained at different times, a value of a direction of the object in the form of a direction of movement of the object. The method according to any one of claims 1 to 25, comprising: 27. Apparatus for performing the method according to any one of claims 1 to 26.